{"id":20603925,"url":"https://github.com/openmlsys/openmlsys-cuda","last_synced_at":"2025-10-08T23:24:16.397Z","repository":{"id":40613770,"uuid":"484397673","full_name":"openmlsys/openmlsys-cuda","owner":"openmlsys","description":"Tutorials for writing high-performance GPU operators in AI frameworks.","archived":false,"fork":false,"pushed_at":"2023-08-12T14:47:36.000Z","size":86,"stargazers_count":130,"open_issues_count":2,"forks_count":16,"subscribers_count":2,"default_branch":"main","last_synced_at":"2025-04-22T03:30:55.059Z","etag":null,"topics":["cuda","gpu","machine-learning"],"latest_commit_sha":null,"homepage":"","language":"Cuda","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":null,"status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/openmlsys.png","metadata":{"files":{"readme":"README.md","changelog":null,"contributing":null,"funding":null,"license":null,"code_of_conduct":null,"threat_model":null,"audit":null,"citation":null,"codeowners":null,"security":null,"support":null,"governance":null,"roadmap":null,"authors":null,"dei":null,"publiccode":null,"codemeta":null}},"created_at":"2022-04-22T10:50:11.000Z","updated_at":"2025-04-15T15:36:49.000Z","dependencies_parsed_at":"2024-11-16T09:19:36.226Z","dependency_job_id":"4c73327e-0971-406f-93e4-98706683df0d","html_url":"https://github.com/openmlsys/openmlsys-cuda","commit_stats":null,"previous_names":[],"tags_count":0,"template":false,"template_full_name":null,"purl":"pkg:github/openmlsys/openmlsys-cuda","repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/openmlsys%2Fopenmlsys-cuda","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/openmlsys%2Fopenmlsys-cuda/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/openmlsys%2Fopenmlsys-cuda/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/openmlsys%2Fopenmlsys-cuda/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/openmlsys","download_url":"https://codeload.github.com/openmlsys/openmlsys-cuda/tar.gz/refs/heads/main","sbom_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/openmlsys%2Fopenmlsys-cuda/sbom","scorecard":null,"host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":279000736,"owners_count":26082862,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2022-07-04T15:15:14.044Z","status":"online","status_checked_at":"2025-10-08T02:00:06.501Z","response_time":56,"last_error":null,"robots_txt_status":"success","robots_txt_updated_at":"2025-07-24T06:49:26.215Z","robots_txt_url":"https://github.com/robots.txt","online":true,"can_crawl_api":true,"host_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub","repositories_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories","repository_names_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repository_names","owners_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners"}},"keywords":["cuda","gpu","machine-learning"],"created_at":"2024-11-16T09:19:28.878Z","updated_at":"2025-10-08T23:24:16.372Z","avatar_url":"https://github.com/openmlsys.png","language":"Cuda","funding_links":[],"categories":[],"sub_categories":[],"readme":"# openmlsys-cuda\n\nExamples for beginners to write your own high-performance AI operators. We introduced optimizations tricks like using shared memory and pipeline rearrangement to maximize the throughput. We also provided an example for using CUTLASS to implement an FC + ReLU fused operator.\n\n## Dependencies\n\n- Eigen: CPU linear algebra template library\n- OpenMP: Enable multi-threads acceleration on CPU\n- CUDA toolkit: Compile GPU kernels and analyse GPU executions\n- Gflags: Commandline flags library released by Google\n- CUTLASS: GPU GEMM template library\n\n### Installation Hints\n\n- Eigen: Use package manager, e.g. `apt install libeigen3-dev`, or download from\n  the [official website](https://eigen.tuxfamily.org/) and build from source.\n- OpenMP: Most time the compilers have already integrated with OpenMP. If your compiler does not support OpenMP,\n  try `apt install libgomp-dev` or `apt install libomp-dev` for GCC or Clang separately.\n- CUDA toolkit: It's recommended to install following\n  the [official instructions](https://developer.nvidia.com/cuda-toolkit).\n- Gflags: Use package manager, e.g. `apt install libgflags-dev`, or download from\n  the [official website](https://gflags.github.io/gflags/) and build from source.\n- CUTLASS: We have registered it to our git module, so you do not have to install by yourself.\n\n## Compilation\n\nOnce you have installed the dependencies, you can use the following instruction to compile the project:\n\n```bash\ngit clone git@github.com:openmlsys/openmlsys-cuda.git\ncd openmlsys-cuda\ngit submodule init \u0026\u0026 git submodule sync\nmkdir build \u0026\u0026 cd build\ncmake ..\nmake -j4\n```\n\n## Examples\n\n- `first_attempt`: The naive implementation\n- `gemm`: Collection of implementations using different optimization tricks\n- `fc_relu`: Example for fusing FC and ReLU by using CUTLASS\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fopenmlsys%2Fopenmlsys-cuda","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fopenmlsys%2Fopenmlsys-cuda","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fopenmlsys%2Fopenmlsys-cuda/lists"}